fantel
bofreq-geng-fantel-09

Description

The fantel BOF (Fast Notification for Traffic Engineering and Load Balancing) aims to discuss notification mechanisms in adaptive/perceptive routing to address dynamic traffic protection, congestion, and performance issues. It focuses on the requirements of fast notification for efficient routing decisions based on real-time network sensing and conditions, facilitating more responsive restoration of connectivity, traffic engineering, and load balancing.

The purpose of this BoF is to highlight the requirements and to bring together a community of interested parties to see whether is agreement on the use cases and requirements, and to see whether there is a group of people who want to work on technology solutions.

Required Details

• Status: non-WG Forming
• Responsible AD: James Guichard
• BOF proponents: Xuesong Geng <gengxuesong@huawei.com>, Tianran Zhou <zhoutianran@huawei.com>, Dan Li <tolidan@tsinghua.edu.cn>, PengFei Huo <huopengfei@bytedance.com>, Weiqiang Cheng <chengweiqiang@chinamobile.com>, Yongqing Zhu <zhuyq8@chinatelecom.cn>, Chang Cao <caoc15@chinaunicom.cn>
• Number of people expected to attend: 100
• Length of session (1 or 2 hours): 2 hours
• Conflicts (whole Areas and/or WGs)
• RTG Area
• Conflicts (whole Areas and/or WGs)
• Chair Conflicts: RTGWG/IDR/LSR
• Technology Overlap: Traffic Engineering, Load Balancing, Failure Protection, Congestion Avoidance
• Key Participant Conflict: Jeffery Zhang, Jie Dong

Information for IAB/IESG

Ongoing efforts in adaptive and perceptive routing are being explored in WGs like RTGWG, LSR, and IDR. These efforts have been substantially triggered by discussion of routing within Data Centres (DCs), especially those with highly-redundant topologies, but wider applicability in scenarios (such as DC inter-connect) has also attracted interest.

However, these efforts are scattered across different WGs without sufficient alignment and there appears to be no consensus on terminology, solutions, or standard requirements in this topic. These different viewpoints have been visible in discussions in RTGWG and a series of side-meetings culminating in a successful perceptive/adaptive routing side-meeting at IETF-121. This all underscores the need for further discussion and alignment within a BOF.

These works are closely related to the fast notification mechanism we plan to discuss in the BoF, to determine:

• Which (if any) modifications to existing protocols or practices are required; Which (if any) entirely new protocols or practices are required:
  Several existing documents discuss potential solutions for fast notification, which could be implemented using IGP or BGP extensions. Fantel could be realized by modifying existing protocols or, if necessary, introducing a new protocol in different deployment scenarios.

There are some private implementations related to this work within data centers. Additionally, several tests are currently under discussion, based on existing documents, to explore and validate potential solutions for fast notification.

Draft Agenda

• Administrivia
• Introduction and purpose of the BoF
• Fantel Use Case
• Existing Tools and Gap Analysis
• Potential Solutions
• Open Discussion

Links to the mailing list, draft charter if any, relevant Internet-Drafts, etc.

• Mailing List: https://mailman3.ietf.org/mailman3/lists/fantel.ietf.org/
• Relevant Internet-Drafts:
  • Scenario:
    • https://datatracker.ietf.org/doc/draft-wh-rtgwg-adaptive-routing-arn/
    • https://datatracker.ietf.org/doc/draft-cheng-rtgwg-adaptive-routing-framework/
    • https://datatracker.ietf.org/doc/draft-liu-rtgwg-path-aware-remote-protection/
  • Information Model
    • https://datatracker.ietf.org/doc/draft-zhou-rtgwg-perceptive-routing-information/
  • State of the art
    • https://datatracker.ietf.org/doc/html/draft-dong-fantel-state-of-art